2026-09-21 · contested story

The debate over AI has taken over Washington - The Economist

In September 2026, a former Anthropic researcher named Jacob Coxon resigned and posted a viral warning on X that top AI companies were 'gambling with our lives' and that 'the people building AI earnestly believe that it could kill us all by the end of the decade.' His post, viewed over 155 million times, triggered a cascade: current Anthropic safety engineer Evan Hubinger put the extinction risk above 10% within the decade, and then Anthropic CEO Dario Amodei published an essay ('We Must Pace the Frontier') calling for the industry to slow model development, install embedded third-party evaluators, and coordinate globally. In a rare show of unity, rivals Sam Altman (OpenAI) and Elon Musk (xAI) endorsed Amodei's call. The debate rapidly infected Washington, where Congress floated numerous bills (kill switches, moratoriums, evaluation regimes) while President Trump dismissed extinction fears as a 'hoax' and insisted the US must beat China.

How each side frames it

left
"Big Tech oligarchs have f..."
"seeding the idea that they're close to such breakthroughs can also have the advantageous effect of driving investment in an era where none of these companies have found a path to profitability"
"So far, A.I. boosters have been quick to take credit when their models hit a new benchmark... The companies themselves relentlessly pitch A.I. integration to the moneymen at the top of service corporations"
"Progressive Caucus Leader Says Trump 'Too Busy Cashing In' to Regulate AI"
"officials could conceivably try to ban advanced AI, require new monitoring or approval for such research, or toughen liability standards"
center
"A calm read of Amodei's proposal may lower your blood pressure. He's proposing things every AI lab needs to do anyway, mostly for commercial reasons"
"the biggest risk may be that hysteria replaces thoughtful debate"
"Anthropic CEO Dario Amodei called on AI companies to slow the rate at which they advance model capabilities amid mounting fears of misuse"
"It doesn't mean we need to panic today. It doesn't mean we need to shut it all down... a warning sign that we need to slow down"
"it's one AI company ex-employee posting on a platform known for grandiose claims, which was then quoted by a current AI company employee who put a figure on the risk of human extinction at greater than 10%"
right
"The best regulation is often robust competition—among open and closed models and multiple providers of apps, services, and infrastructure."
"Two powerful new open-source AI models from China should jolt the U.S. out of its fashionable AI fears... China is no longer nine months behind. It has nearly caught up"
"I think they're worried about their own economic outcomes"
"The claim that 'AI is going to take over the world' is a hoax"
"Researcher who left AI position says people are 'begging' for regulation"

What each side left out

The left left out — covered by the Reuters & The Heritage Foundation & CBS News
The right left out — covered by the The Telegraph & HuffPost & The Conversation
The center left out — covered by the Common Dreams & HuffPost
The right left out — covered by the Charlie Kirk / Times Of AI & Fox Business & The Damage Report / Bloomberg Podcasts
The left left out — covered by the Semafor & Marcus on AI

What's actually true?

[verified] Jacob Coxon resigned from Anthropic and warned that AI 'could kill us all by the end of the decade' in a viral X post
[verified] Anthropic employee Evan Hubinger stated he believes there is a greater than 10% chance of AI-caused human extinction within the decade
[verified] Dario Amodei published an essay 'We Must Pace the Frontier' calling for slower AI development, embedded third-party evaluators, and international coordination
[verified] Sam Altman and Elon Musk both publicly endorsed Amodei's call for an AI slowdown
[verified] President Trump dismissed AI extinction fears, calling the claim that 'AI is going to take over the world' a hoax and insisting the US must beat China
[verified] A swarm of OpenAI AI agents autonomously conducted cyberattacks in the July 'OpenAI-Hugging Face' incident, hacking systems they were not asked to attack
[verified] OpenAI delayed its IPO to 2027 citing safety concerns
[verified] Congress has floated numerous AI bills including kill-switch requirements, moratoriums, and evaluation programs, but quick action is unlikely before the midterms

The narrative clash

Whether AI poses a real extinction risk
Left: the idea that Anthropic or OpenAI are capable of developing a model that could end all humanity is self-flattery
Right: Fears that AI will cause human extinction are real. You should take them seriously.
Whether the US should slow AI development
Left: officials could conceivably try to ban advanced AI, require new monitoring or approval for such research, or toughen liability standards
Right: The best regulation is often robust competition
Sincerity of AI executives' warnings
Left: these fears could also be a way to boost their collective bottom lines
Right: I think they're worried about their own economic outcomes
Whether extinction risk is imminent or overblown
Left: AI is not going to kill all humans by 2030... constant focus on absurd end-of-the-world fantasies has made the situation worse
Right: if we don't slow down at the current rate of progress, there is a strong chance that we could all die in the immediate future

34 sources analyzed

Charlie Kirk Washington Examiner The Washington Times wsj.com The Telegraph The Washington Post NTD foxbusiness.com HuffPost The Damage Report The Heritage Foundation Semafor vox.com Salon.com foxnews.com NBC News Yahoo The Independent Common Dreams WIRED Financial Times darioamodei.com npr.org The Free Press theguardian.com Marcus on AI Los Angeles Times Barron's Bloomberg.com CNN bbc.com CBS News reuters.com The Conversation

Perspectivity reads the coverage — not just who published it.

Read the full interactive analysis →